Method for locating road objects

Abstract

A method and device for locating road objects based on a plurality of routes transmitted by at least one vehicle driven on a road network. A route transmitted by a vehicle including a plurality of successive vehicle locations acquired during a driving session. At least one road object detected during the driving session by a sensor of the vehicle, the road object associated with a location of the vehicle at the time of detection. The method selects routes having at least one calibration road object in a list of calibration road objects. The list is initialized with at least one road object with a known location. For each selected route, at least one readjustment parameter is computed. The computed readjustment parameter applied to the objects in the route to obtain a calibrated route. The list of calibration objects being updated with the locations of road objects in the readjusted route.

The invention claimed is: 
     
       1. A method for locating road objects based on a plurality of routes transmitted by at least one vehicle being driven on a road network, the method comprising the following steps:
 initializing a list of calibration road objects with at least one calibration object associated with an actual location that is known, wherein each calibration object in the list of calibration road objects is associated with a respective actual location that is known; 
 selecting routes, among the plurality of routes, comprising at least one road object corresponding to the at least one calibration object in the list of calibration road objects, wherein each of the selected routes is transmitted by a respective vehicle and comprises:
 a plurality of successive vehicle locations acquired during a driving session, and 
 the at least one road object detected during the driving session by a sensor of the vehicle, the at least one road object being associated with a location of the vehicle at a time of its detection; 
 
 for each of the selected routes,
 computing at least one readjustment parameter representative of a discrepancy between the location associated with the at least one road object in the route, and the actual location associated with the at least one calibration object in the list of calibration road objects, 
 applying the at least one computed readjustment parameter to locations of all road objects included in the route to obtain a calibrated route, and 
 updating the list of calibration road objects with the locations of all road objects included in the calibrated route; and 
 
 generating a high-resolution map of the road network based on the calibrated routes and the updated list of calibration road objects, 
 the steps of selecting, computing, applying and updating being repeated when the at least one computed readjustment parameter is above a particular threshold. 
 
     
     
       2. The method as claimed in  claim 1 , wherein the step of computing the readjustment parameter comprises:
 computing a first discrepancy between a location associated with a first object in the route, and a location associated with said first object in the list of calibration road objects; and 
 computing at least a second discrepancy between a location associated with a second object in the route, and a location associated with said second object in the list of calibration road objects. 
 
     
     
       3. The method as claimed in  claim 2 , wherein the first and second discrepancies are weighted by confidence indices associated with the locations of the first and second objects in the list of calibration road objects, respectively, a confidence index of the confidence indices being inversely proportional to a number of iterations of the steps of selecting, computing, applying and updating that preceded an update of the corresponding object in the list of calibration road objects. 
     
     
       4. The method as claimed in  claim 3 , wherein the step of computing the readjustment parameter further comprises computing an average of the first and second discrepancies. 
     
     
       5. The method as claimed in  claim 2 , wherein the step of computing the readjustment parameter further comprises computing an average of the first and second discrepancies. 
     
     
       6. The method as claimed in  claim 5 , wherein the computed average is an average weighted by confidence indices associated with the locations of the first and second objects in the list of calibration road objects, respectively. 
     
     
       7. The method as claimed in  claim 1 , wherein, when a route among the selected routes comprises at least first and second road objects corresponding to first and second calibration objects, respectively, in the list of calibration road objects for which first and second readjustment parameters are respectively computed, a third readjustment parameter is estimated for a third road object located temporally in the route between the first and second calibration objects, the estimation being carried out via a regression based on the first and second readjustment parameters computed for the first and second calibration objects and on a respective time of detection of the first, second and third road objects. 
     
     
       8. The method as claimed in  claim 7 , wherein the regression is weighted by confidence indices associated with the first and second calibration objects. 
     
     
       9. A non-transitory processor-readable data medium storing a computer program comprising instructions for executing the steps of the method of  claim 1 . 
     
     
       10. A device for locating road objects based on a plurality of routes transmitted by at least one vehicle being driven on a road network,
 the device comprising a processor and a memory storing computer program instructions that, when executed by the processor, cause the processor to implement steps comprising:
 initializing a list of calibration road objects with at least one calibration object associated with an actual location that is known, wherein each calibration object in the list of calibration road objects is associated with a respective actual location that is known; 
 selecting routes, among the plurality of routes, comprising at least one road object corresponding to the at least one calibration object in the list of calibration road objects, wherein each of the selected routes is transmitted by a respective vehicle and comprises:
 a plurality of successive vehicle locations acquired during a driving session, and 
 the at least one road object detected during the driving session by a sensor of the vehicle, the at least one road object being associated with a location of the vehicle at a time of its detection; 
 
 for each of the selected routes,
 computing at least one readjustment parameter representative of a discrepancy between the location associated with the at least one road object in the route, and the actual location associated with the at least one calibration object in the list of calibration road objects, 
 applying the at least one computed readjustment parameter to locations of all road objects included in the route to obtain a calibrated route, and 
 updating the list of calibration road objects with the locations of all road objects included in the calibrated route; and 
 
 generating a high-resolution map of the road network based on the calibrated routes and the updated list of calibration road objects, 
 
 the steps of selecting, computing, applying and updating being repeated when the at least one computed readjustment parameter is above a particular threshold. 
 
     
     
       11. A server comprising the device claimed in  claim 10 .